The specificity of a previously developed 57-criteria/32-point QRS scoring system for estimating myocardial infarct (MI) size is evaluated in an extensive control population and the method of application of this system for determining a QRS score from a standard 12-lead electrocardiogram is described. Points are accumulated from Q- and R- wave durations, R- and S-wave amplitudes, R/Q- or R/S-amplitude ratios and the presence of R-wave notching, with each point representing approximately 3% of the left ventricle. The subjects were selected because of the minimal likelihood of their having had myocardial infarcts or other sources of QRS modification. There were 500 consecutively selected normal Caucasian subjects, aged 20 to 69 years, with 50 women and 50 men in each of the 5 decades. Specificity for the 57 individual criteria ranged from 89 to 100 %. Fifty-one criteria met the required standard of at least 95 % specificity; of the 6 that failed, 3 were successfully modified to achieve this standard and 3 were eliminated. In the resultant 54-criteria/32-point complete system, the total population, as well as both women and men, required more than 3 points to attain at least 95% specificity. Subjects in each of the 5 decades met the specificity standard either at or below the level of more than 3 points. The point score at which 95% or greater specificity was attained for the 10 age/sex subsets varied. The 60- to 69-year-old men were the least specific subset, achieving this standard at more than 4 points, while the 40- to 49-year-old men were the most specific, requiring only a score of more than 1 point. Thus, the complete scoring system is specific overall in the adult Caucasian population studied here. However, because the specificity of many criteria were dependent on age and sex, this system must be further evaluated by consideration of a more comprehensive database. © 1985.